SOLVING THE UNITED KINGDOM'S PRODUCTIVITY PUZZLE IN A DIGITAL AGE - MCKINSEY
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
SOLVING THE
UNITED KINGDOM’S
PRODUCTIVITY PUZZLE
IN A DIGITAL AGE
DISCUSSION PAPER
SEPTEMBER 2018
Jacques Bughin | Brussels
Jonathan Dimson | London
Vivian Hunt | London
Tera Allas | London
Mekala Krishnan | Boston
Jan Mischke | Zurich
Louis Chambers | London
Marc Canal | MadridSince its founding in 1990, the McKinsey Global Institute (MGI) has sought to develop a deeper understanding of the evolving global economy. As the business and economics research arm of McKinsey & Company, MGI aims to provide leaders in the commercial, public, and social sectors with the facts and insights on which to base management and policy decisions. MGI research combines the disciplines of economics and management, employing the analytical tools of economics with the insights of business leaders. Our “micro-to-macro” methodology examines microeconomic industry trends to better understand the broad macroeconomic forces affecting business strategy and public policy. MGI’s in-depth reports have covered more than 20 countries and 30 industries. Current research focuses on six themes: productivity and growth, natural resources, labour markets, the evolution of global financial markets, the economic impact of technology and innovation, and urbanisation. Recent reports have assessed the digital economy, the impact of AI and automation on employment, income inequality, the productivity puzzle, the economic benefits of tackling gender inequality, a new era of global competition, Chinese innovation, and digital and financial globalisation. MGI is led by three McKinsey & Company senior partners: Jacques Bughin, Jonathan Woetzel, and James Manyika, who also serves as the chairman of MGI. Michael Chui, Susan Lund, Anu Madgavkar, Jan Mischke, Sree Ramaswamy, and Jaana Remes are MGI partners, and Mekala Krishnan and Jeongmin Seong are MGI senior fellows. Project teams are led by the MGI partners and a group of senior fellows, and include consultants from McKinsey offices around the world. These teams draw on McKinsey’s global network of partners and industry and management experts. Advice and input to MGI research are provided by the MGI Council, members of which are also involved in MGI’s research. MGI Council members are drawn from around the world and from various sectors and include Andrés Cadena, Sandrine Devillard, Richard Dobbs, Tarek Elmasry, Katy George, Rajat Gupta, Eric Hazan, Eric Labaye, Acha Leke, Scott Nyquist, Gary Pinkus, Sven Smit, Oliver Tonby, and Eckart Windhagen. In addition, leading economists, including Nobel laureates, act as advisers to MGI research. The partners of McKinsey fund MGI’s research; it is not commissioned by any business, government, or other institution. For further information about MGI and to download reports, please visit www.mckinsey.com/mgi. IN COLLABORATION WITH MCKINSEY & COMPANY McKinsey & Company is a global management consulting firm, deeply committed to helping institutions in the private, public, and social sectors achieve lasting success. For more than eight decades, our primary objective has been to serve as our clients’ most trusted external adviser. With consultants in 124 cities and 63 countries, we bring unparalleled expertise to clients across all relevant industries and functions, anywhere in the world. We work closely with teams at all levels of an organisation to shape winning strategies, mobilise for change, build capabilities, and drive successful transformation and lasting improvements. McKinsey’s UK office, based in London, was the firm’s first overseas office, and has a long history of serving both British and global private, public, and social organisations. It actively contributes to the public debate on addressing the United Kingdom’s productivity challenges and promoting economic prosperity and growth. Copyright © McKinsey & Company 2018
CONTENTS
Acknowledgments
In brief
Page 2
Introduction
Page 4
1. An international comparison
Page 7
2. Reasons for the UK slowdown
Page 13
3. Steps to accelerate UK productivity growth
Page 29
ACKNOWLEDGMENTS
This paper is our attempt to shed light on the productivity Roberto Migliorini, Daniel Mikkelsen, Jaana Remes,
puzzle for the United Kingdom. We have tried to provide Maciej Szymanowski, Yoshi Takanuki, and Richard Verity.
a useful understanding of why productivity growth has
slowed sharply in the United Kingdom in recent years. In For their invaluable discussions and insights, many
addition, we have tried to create a framework for how to thanks go to: Tommaso Aquilante, senior economist
accelerate productivity growth in the country in the future. at the Bank of England; Martin N. Baily, Bernard L.
Schwartz Chair in Economic Policy Development and
This paper was a joint project between McKinsey senior fellow, economic studies, Center on Regulation
& Company’s UK and Ireland office and MGI. The and Markets at the Brookings Institution and academic
research was led by Tera Allas, senior fellow at the adviser to MGI; Diane Coyle, Bennett Professor of Public
McKinsey Center for Government, and Jonathan Dimson Policy, fellow, Churchill College, University of Cambridge;
and Vivian Hunt, both senior partners of McKinsey Richard Davies, Centre for Economic Performance,
& Company, together with Jacques Bughin, a London School of Economics; Philip Duffy, director,
director of MGI and senior partner of McKinsey Enterprise and Growth, HM Treasury; Andy Haldane, chief
& Company, Jan Mischke, an MGI partner, and economist at the Bank of England; Richard Heys, deputy
Mekala Krishnan, a senior fellow at MGI. The project chief economist, Office for National Statistics; Lee Hopley,
team was led by Louis Chambers and Marc Canal chief economist, EEF, the manufacturers’ organisation;
and included Samuel Cudre, Kimberley Moran, and Dr Dan Mawson, head of the Economy Team, BEIS;
Naba Salman. This paper was edited and produced Rebecca Riley, director, Economic Statistics Centre of
by senior editor Anna Bernasek, editorial production Excellence and fellow, National Institute of Economic
manager Julie Philpot, digital editor Lauren Meling, and Social Research; Patrick Schneider, economist at
senior graphic designer Marisa Carder, and graphic the Bank of England; and Martin Weale, professor of
design specialist Margo Shimasaki. John Cheetham, economics at King’s College London.
Ian Gleeson, and Nicola Montenegri together with
Nienke Beuwer managed dissemination and publicity. We are grateful for all the input we have received,
but the final paper is ours, and all errors are our own.
We are very grateful for all the help we received from We welcome your comments on this research at
McKinsey and MGI colleagues including Richard Dobbs, MGI@mckinsey.com.IN BRIEF SOLVING THE UNITED KINGDOM’S PRODUCTIVITY PUZZLE IN A DIGITAL AGE The United Kingdom went into the financial crisis with one of the lowest labour productivity levels among European peers and emerged with the steepest decline in productivity growth. While productivity growth has recovered slightly since 2014, it remains near historic lows. Developing headwinds, such as demographic shifts and uncertainty surrounding Brexit, mean that improving productivity will be more important than ever for raising living standards. We identify key reasons for the United Kingdom’s recent weak productivity performance by analysing cross-country, regional, and sectoral patterns as well as other decompositions of aggregate statistics, and outline steps policy makers and firms can take to promote growth. The United Kingdom’s labour productivity growth slowdown was broader-based than France, Germany, and Spain, occurring across all regions and 83 percent of sectors. Despite this, the financial and manufacturing sectors played outsize roles, as did a drop in total factor productivity (TFP) growth. We identify four main reasons for these patterns. The financial sector experienced a boom ahead of the crisis and a bust in the aftermath, accounting for about 20 percent of the productivity growth slowdown in the United Kingdom and a third of the decline in TFP growth. While the United States also experienced a boom/bust cycle in finance, the impact was much smaller, only about 10 percent. Growth in loans and deposits contracted sharply in the United Kingdom after the financial crisis while the fixed nature of many inputs meant that hours worked barely changed. The United Kingdom stands out as having the strongest growth in hours worked in our European sample after the crisis, three times the average rate of our sample. The increase in hours reflected not merely a rebound from the crisis but additional hiring, with two million more people, especially the young and old, employed between 2010 and 2015. Policy changes to apprenticeships, university fees, and pensions may have influenced labour supply dynamics. Low wage growth reinforced hiring ahead of capital investment, especially given the heightened economic uncertainty. UK investment was the lowest in our sample of advanced economies going into the crisis and fell further in the aftermath. The decline was mainly from a reduction in equipment and structures investment, while investment in intangibles such as software and R&D increased slightly. Weak equipment investment has implications for achieving labour savings and played a role in the productivity growth decline in manufacturing. While the United Kingdom ranks highly in broad measures of digital adoption, there are gaps. The country does well in internet access, basic digital skills, and the adoption of cloud computing but poorly in the integration of information systems across the value chain, business process transformations, enterprise digitisation, and robotics. We find opportunities to boost productivity growth from digitisation but they come with adoption barriers, lags, and transition costs, and the benefits have not yet occurred at scale. Across countries we analysed, there is potential for at least 2 percent productivity growth a year over the next ten years. However, capturing that potential in the United Kingdom will take time and is not guaranteed. It will require policy makers and businesses to take decisive action in key areas: skill building for the existing and future workforce and managers; accelerating adoption of digital through better information, access to finance, collaborations, and a favourable policy environment; and promoting additional investment and exports.
WHAT HAPPENED TO UK
PRODUCTIVITY GROWTH?
Productivity growth in the UK averaged
2% before the financial crisis
(2000–05), but only
The financial and
manufacturing sectors
0.2% after (2010–15)
accounted for nearly
Hours worked growth was
50% of the
aggregate
slowdown 3x that of comparable countries
in 2010–15, pointing to an
employment puzzle, with
2M more people
employed
Equipment investment as a share UK manufacturing
of value added declined by robotics adoption
20% post- vs pre-crisis,
~20%
lags at
23%
of which
50%
came from
of UK companies
cited various
of Germany’s rate
forms of
manufacturing
uncertainty as a
main reason for
not investing
Yet advanced economies, including the UK,
could boost annual productivity growth to
2%+
with supportive actionINTRODUCTION
Declining labour productivity growth characterized many advanced economies after a boom
in the 1960s, but since the mid-2000s that decline has accelerated. Against that backdrop,
the United Kingdom stands out as one of the worst productivity performers among its
peers. Its absolute level of productivity has persistently ranked towards the bottom of
a sample of advanced economies. Moreover, in the aftermath of the crisis, the United
Kingdom, along with the United States, recorded one of the lowest productivity growth rates
and steepest declines in productivity growth compared to those same countries, falling by
90 percent.1 Between 2010 and 2015, UK productivity growth flatlined at 0.2 percent a year,
far below its long-term average of 2.4 percent from 1970 to 2007. Productivity growth is
important for all advanced economies as they navigate potential economic headwinds, such
as an aging population and an ongoing shift to low-productivity services like healthcare and
education, but particularly for the United Kingdom with an uncertain outlook for trade and
investment after Brexit.2
Several explanations have been offered for the United Kingdom’s poor productivity
performance. These include low capital investment after the financial crisis and Brexit
referendum, poor reallocation of capital to the most productive businesses, a productivity
growth slowdown in key sectors such as finance and telecommunications after a pre-
crisis boom, a slowdown in the rate of innovation and diffusion, and mismeasurement of
output growth (see Box 1, “The role of mismeasurement in the UK productivity growth
slowdown”).3 In this paper, we bring together a comprehensive evidence base, drawing from
international comparisons and country-specific analyses, to highlight the distinct features
of the United Kingdom’s productivity growth slowdown. The resulting insights provide a
fresh perspective on a stubborn question and shed light on what can be done to restart UK
productivity growth.
Our analysis builds on recent productivity research at the McKinsey Global Institute (MGI),
McKinsey & Company’s UK and Ireland office, and the McKinsey Center for Government.
In 2017, MGI launched a research effort investigating the productivity growth slowdown
since the mid-2000s across major European countries including the United Kingdom, and
the United States. That research culminated in the publication of Solving the productivity
puzzle: The role of demand and the promise of digitization.4 In addition, recent research by
1
Our sample of five advanced economies includes France, Germany, Spain, the United Kingdom, and the
United States. Note that this discussion paper focuses on explanations for and policy implications of the
productivity growth slowdown between 2000–05 and 2010–15 using primarily sector-level evidence.
We therefore do not elaborate on the underlying reasons for the United Kingdom’s low absolute level of
productivity and draw attention to these only where they are also relevant to the recent productivity growth
slowdown. As a result, the paper does not cover in detail, for example, issues of demand, inequality,
workforce diversity, infrastructure, supply chains, skills mismatches, management practices, regional
productivity differences, innovation diffusion, or firm-level evidence (e.g., the role of small and medium-size
enterprises). Of course, some of the reasons for low productivity levels do point to potential actions for
boosting future productivity growth, and we capture the most important of these in the recommendations
section of the paper.
2
The measurement of labour productivity in health and education is problematic and is unlikely to capture the
full value of these services; however, as measured in national accounts, their relative level of productivity is low.
3
See, for example, UK productivity trends: How are they influenced by financial services? TheCityUK
Independent Economist Group, May 2013; Peter Goodridge, Jonathan Haskel, and Gavin Wallis, “Accounting
for the UK productivity puzzle: A decomposition and predictions”, Economica, December 2016; João Paulo
Pessoa and John Van Reenen, “The UK productivity and jobs puzzle: Does the answer lie in wage flexibility?”
Economic Journal, May 2014, Volume 124, Issue 576, pp. 433–52; Alina Barnett, Adrian Chiu, Jeremy
Franklin, and Maria Sebastia-Barriel, The productivity puzzle: A firm-level investigation into employment
behaviour and resource allocation over the crisis, Bank of England working paper number 495, April 2014;
“Productivity puzzles”, speech by Andy Haldane, Bank of England, at London School of Economics, March
2017; Diane Coyle, Do it yourself digital: The production boundary and the productivity puzzle, Economic
Statistics Centre of Excellence discussion paper number 2017-01, June 2017; Diane Coyle, “Commentary:
Modernising economic statistics: Why it matters”, National Institute Economic Review, November 2015,
Volume 234, Issue 1.
4
Solving the productivity puzzle: The role of demand and the promise of digitization, McKinsey Global Institute,
February 2018.McKinsey’s UK and Ireland office has highlighted the importance of management skills,
innovation diffusion, and regional variations to understanding the productivity puzzle, while
the McKinsey Center for Government has identified vast opportunities from unlocking
productivity growth in the public sector.5
This paper extends the analysis and findings of MGI’s Solving the productivity puzzle to
provide a deeper understanding of the United Kingdom’s most recent period of reduced
productivity growth following the financial crisis. The MGI report found that the waning
of a productivity boom—fuelled by information and communications technology (ICT),
outsourcing and restructuring that began in the 1990s—and financial crisis aftereffects,
including weak demand and uncertainty, dragged down productivity growth by
1.9 percentage points on average, from 2.4 percent to 0.5 percent, across the United
States and Western Europe for the period 2010 to 2014 compared to the period 2000 to
2004.6 It also found that digitisation was occurring across these countries with significant
potential to reignite productivity growth, but the benefits had not yet materialised at scale.
The slow impact of digitisation on productivity growth is due to adoption barriers and lag
effects as well as transition costs. The report concluded that as financial crisis aftereffects
continue to recede and more companies incorporate digital solutions, productivity growth
has the potential to recover across countries. It calculated that the productivity growth
potential from both digital and nondigital opportunities could be at least 2 percent per year
over the next decade. However, realising this opportunity would take time, was no means
guaranteed, and would require a combination of supportive government policy and action
by companies. In particular, the report highlighted that capturing the productivity potential of
advanced economies will require a dual focus that promotes sustained demand growth and
digital diffusion.
While these trends hold true for the United Kingdom, there are important differences,
which we highlight in this paper. In Chapter 1, we describe what has happened to UK
productivity growth in the aftermath of the financial crisis and put that in the context of
other advanced economies. In Chapter 2, we identify the reasons for the United Kingdom’s
recent productivity growth slowdown: the role of the financial sector, a focus on adding
labour over investment, and gaps in digital adoption, despite overall good digital maturity.
Then, in Chapter 3, we outline steps policy makers and companies can take to promote
productivity growth.
5
Productivity: The route to Brexit success, McKinsey & Company, December 2016; Government productivity:
Unlocking the $3.5 trillion opportunity, McKinsey Center for Government, April 2017.
6
In this paper, we update the time periods to reflect the latest data available (KLEMS 2017 release), and
compare productivity growth in the period 2010–15 to that in the period 2000–2005. In our original analysis,
we compared the turn of the century (2000–04)—a five-year period before the start of the recent productivity
growth slowdown—with the post-recession years (2010–14), a somewhat stable period a decade later
(though encompassing the double-dip recession in Europe). While we are aware that choosing specific
years involves a degree of arbitrariness, after assessing the pros and cons of multiple different periods,
we concluded that concentrating on the period following the crisis allowed us to isolate different factors at
the sector level across many different countries more easily. Yet, as explained in footnote 1 above, it also
means that in this paper, we do not address in detail the long-term factors behind the United Kingdom’s low
productivity levels such as infrastructure gaps or poor management practices.
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 5Box 1. The role of mismeasurement in the UK productivity growth slowdown
The measurement of productivity raises many difficult challenges. Output is hard to measure
in services, particularly in public services such as health and education. Quality improvements
in many areas, especially tech and software, are hard to adjust for accurately. New consumer
services, often provided free of charge—such as mobile GPS, search engines, a host of
smartphone-based applications, and cloud-based services—have contributed to consumer
welfare in ways we are currently not measuring. Nondigital issues, such as globalisation of
value chains and profits shifting overseas, and investment in intangibles also contribute to the
measurement challenge. These issues indicate that actual productivity is probably higher than
measured productivity.
While the amount of mismeasurement is likely to be significant, we find it is not sufficient to
explain the full extent of the recent productivity growth slowdown. We identify a broad-based
productivity growth slowdown across sectors of the economy, indicating that sector-specific
mismeasurement issues cannot fully explain the economy-wide slowdown. We also identify
factors that explain the productivity slowdown via clear non-measurement-related channels, as
described later in this paper.
Various researchers have attempted to size the portion of the productivity growth shortfall
attributable to mismeasurement. A review of the literature suggests that mismeasurement
is likely to account for at most a third of the recent slowdown.1 While much of this literature
has focused on the United States, we believe that it provides a guide for the United Kingdom
and that the magnitude of the effect would be similar. However, while mismeasurement
seems insufficient to explain the economy-wide decline in productivity growth, it can play a
role in explaining the decline in specific sectors. For example, work by Byrne and co-authors
suggests that accounting for mismeasurement could lead to a smaller decline in productivity
growth in the tech sector and a larger decline in other sectors.2 Research into deflators in
the telecom industry finds that they do not include mobile or broadband data, among other
issues.3 Our analysis finds that telecom explains 0.2 percentage point of the 1.9 percentage
point productivity growth decline in the United Kingdom. However, adjusting the telecom
deflator based on the approach suggested in the research means that telecom would not have
contributed to the decline in productivity growth in the United Kingdom. We also discuss the
role of measurement in the finance sector in Chapter 2.
1
See Chad Syverson, Challenges to mismeasurement explanations for the US productivity slowdown, NBER
working paper number 21974, February 2016. Various authors have suggested a similar conclusion. See Nadim
Ahmad, Jennifer Ribarsky, and Marshall Reinsdorf, Can potential mismeasurement of the digital economy explain
the post-crisis slowdown in GDP and productivity growth? OECD Statistics Working Papers, number 2017/09,
July 2017, and Gustavo Adler et al., Gone with the headwinds: Global productivity, IMF staff discussion notes,
number 17/04, April 2017. Ahmad et al., for example, find that if mismeasurement is occurring, it cannot explain
the magnitude of the observed slowdown in GDP or productivity growth. They do caution that this may not be
true for future growth as the size of the digital economy increases. David M. Byrne, John G. Fernald, and Marshall
B. Reinsdorf have assessed the role played by mismeasurement and find no evidence that mismeasurement has
worsened in recent times. Adjusting price deflators for computers, communications and specialised equipment,
semiconductors, and software, as well as including intangibles, they find, could add about 0.2 percentage point to
US labour productivity growth between 2004 and 2014. However, they believe the mismeasurement contribution
from these factors was actually higher—roughly 0.5 percentage point—between 1995 and 2004, because of
the higher share of domestic production of many of these products in this period. See David M. Byrne, John G.
Fernald, and Marshall B. Reinsdorf, Does the United States have a productivity slowdown or a measurement
problem? Federal Reserve Bank of San Francisco, working paper number 2016-03, April 2016. Another estimate
by Nakamura et al. finds that accounting for free digital content through the lens of a production account would
boost productivity growth, but again, not sufficiently to account for the slowdown. Productivity growth would
be higher by 0.07 percentage point between 1995 and 2005, and by about 0.11 percentage point between
2005 and 2015. See Leonard I. Nakamura, Jon D. Samuels, and Rachel H. Soloveichik, Measuring the “free”
digital economy within the GDP and productivity accounts, Federal Reserve Bank of Philadelphia working paper,
October 2017. Other research on mismeasurement includes Philippe Aghion et al., Missing growth from creative
destruction, Federal Reserve Bank of San Francisco working paper number 2017-04, November 2017.
2
See David Byrne, Stephen Oliner, and Daniel Sichel, Prices of high-tech products, mismeasurement, and pace of
innovation, NBER working paper number 23369, April 2017.
3
Mo Abdirahman et al., “A comparison of approaches to deflating telecoms services output”, Economic Statistics
Centre of Excellence discussion paper number 2017-04, December 2017.
6 McKinsey Global Institute Introduction1. AN INTERNATIONAL COMPARISON
The United Kingdom went into the financial crisis with low labour productivity levels
compared to peers.7 The absolute level of productivity for the United Kingdom in 2007
was about 20 percent lower than for Germany and France and in line with Italy, and these
differentials have not changed today. However, this was not always the case. In 1960, UK
productivity was higher than France’s, but over a 50-year period, productivity diverged as
France experienced faster growth in total factor productivity (TFP) and capital investment (see
Box 2, “How does the United Kingdom’s productivity performance compare to France’s?”).
Against this backdrop of low absolute productivity, UK productivity growth has remained
near historic lows since the financial crisis. Indeed, during the crisis, the decline in
productivity growth in the United Kingdom was more severe than in Europe (Exhibit 1).8
Between 2010 and 2014, UK productivity growth averaged –0.2 percent a year. Since 2014,
the productivity picture has improved somewhat, and from 2014 to 2017 productivity growth
averaged 0.9 percent a year.9 Despite this improvement, UK productivity growth remains
below that of European peers such as France and Germany.
Exhibit 1
The United Kingdom experienced one of the largest declines in productivity growth following the crisis, although
productivity growth has recovered somewhat since 2014.
Trend line of labour productivity growth, total economy
% year-on-year
United Kingdom Europe (excl United Kingdom)1 United States
World Great World Great
War I Depression War II Recession
8
7
6
5
4
3
2
1
0
1870
-1 80 90 1900 10 20 30 40 50 60 70 80 90 2000 10 2020
-2
-3
-4
1 Simple average of France, Germany, Italy, Spain, and Sweden.
NOTE: Productivity defined as GDP per hour worked. Calculated using Hodrick Prescott filter. Drawn from similar analysis in Martin Neil Baily and Nicholas
Montalbano, Why is productivity growth so slow? Possible explanations and policy responses, Brookings Institution, September 2016.
SOURCE: Antonin Bergeaud, Gilbert Cette, and Rémy Lecat, “Productivity trends in advanced countries between 1890 and 2012”, The Review of Income and
Wealth, Volume 62, Number 3, pp. 420–444; The Conference Board, March 2018 release (for 2017 only); McKinsey Global Institute analysis
7
When we use the term “productivity”, we refer to labour productivity, while all other types of productivity—for
example, total factor productivity—are spelled out explicitly.
8 ▪ WWI:
Exhibit 1 uses1914-1918
data from Bergeaud, Cette, and Lecat, to allow us to establish a long-term time series. For
▪ Great
further detailsDepression: 1929-1934
on this data, see Antonin Bergeaud, Gilbert Cette, and Rémy Lecat, “Productivity trends in
advanced countries between 1890 and 2012”, Review of Income and Wealth, September 2016, Volume 62,
▪ WWII:
Number 3. As 1939-1945
a result, data in Exhibit 1 do not exactly match data cited elsewhere in this report, which rely on
▪
EU Great
KLEMS (2017Recession:
release). 2007-2009
9
Based on data from EU KLEMS (2017 release) for data up to 2015 and The Conference Board (March 2018)
for 2016–17 data. This is because the EU KLEMS time series terminates in 2015. While there are differences
between the databases, both databases show a post-crisis productivity slowdown in the United Kingdom,
United States and Europe and an uptick in UK productivity from 2014. For our other analysis in this report, we
rely on data from EU KLEMS (2017 release), unless otherwise stated, in order to draw out sector level insights.
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 7Box 2. How does the United Kingdom’s productivity performance compare to France’s?
Today UK labour productivity is roughly 20 percent lower to produce output. Over the period from 1960 to 2016,
than French labour productivity, but this was not always lower TFP growth in the United Kingdom accounted
the case (Exhibit 2). Until the 1960s, UK productivity for slightly more than half of the productivity growth
was higher than France’s, but then French productivity difference with France, while lower investment growth,
growth began to pull away and has steadily outpaced UK measured by growth in capital intensity or capital services
productivity growth ever since, averaging 2.8 percent a per hour worked, accounted for 40 percent.2
year in France versus 2.3 percent in the United Kingdom.1
Despite higher productivity growth in France, GDP per
Interestingly, the size of the economy and population are
capita, or the amount of goods and services produced
nearly identical, and the mix of sectors is similar. What
divided by the entire population, has remained roughly
explains the difference in productivity performance?
in line with the United Kingdom. The reason is that
Looking across sectors, we found that total factor beginning in the 1980s, total hours worked grew faster
productivity (TFP) growth in France over the long run was in the United Kingdom than in France, while population
higher than in the United Kingdom. Often used as a proxy growth was slower. This means that to make up for its
for technological progress and innovation, TFP reflects lagging productivity growth, the United Kingdom has
the efficiency with which inputs including labour, capital, been relying on growth in employment and hours worked
energy, materials, and purchased services are combined to meet its production needs.
Exhibit 2
The United Kingdom’s labour productivity diverged from France’s in the early 1960s and is now 20 percent lower
while GDP per capita has grown in line.
PPP, constant 2016 $ France United Kingdom Compound annual growth rate,
1960–2016 (%)
Labour productivity (GDP per hour worked) GDP per capita
$ per hour $ thousand per person
70 2.8 50 +4% 2.1
60
40
50
-20% 2.3 1.9
40 30
30 20
20
10
10
0 0
1960 70 80 90 2000 10 2016 1960 70 80 90 2000 10 2016
SOURCE: The Conference Board data (March 2018 release); McKinsey Global Institute analysis
1
The gap reached a maximum of 26 percent in 1990, at which point the United Kingdom began to close the gap, until the financial crisis, when
the gap increased again.
2
Capital intensity is measured as capital services per hour and indicates access to machinery, tools, and equipment. This analysis is based on
data from The Conference Board (March 2018 release).
THE UK PRODUCTIVITY GROWTH DECLINE HAS BEEN EVEN
BROADER-BASED THAN EUROPEAN PEERS
The United Kingdom’s productivity growth slowdown was broad-based across regions
and sectors. Every single UK region saw a productivity growth slowdown relative to the
period before the crisis. This suggests that, even though productivity levels between regions
and local areas in the United Kingdom are very different, the underlying reasons for the
Box 2productivity slowdown were common across geographies.10
10
Unlocking regional growth, Confederation of British Industry, December 2016. The four main reasons for
persistent differences in productivity levels between UK regions and local areas identified in this research were:
education and skills, transport links, management practices, and exports. Differential changes between the
2000–05 period and 2010–15 period in these structural factors are likely to have been relatively small due to
their long-term nature.
8 McKinsey Global Institute 1. An international comparisonThe decline was also broad-based across sectors, with 83 percent (24 of 29 sectors)
experiencing a productivity growth slowdown (Exhibit 3). While this was in line with the
United States, the slowdown across sectors in the United Kingdom is broader-based
compared to its European peers.
Exhibit 3
In the United Kingdom, every region and 83 percent of sectors experienced a productivity growth decline after
the crisis relative to pre-crisis rates.
Productivity growth rate difference between 2010–15 and 2000–05
Percentage points
0 to -0.5
-0.5 to -1.0
-1.0 to -1.5
-1.5 to 2.0
Scotland
-2.0 to -2.5
< -2.5
North East
Northern Ireland
Yorkshire and Humberside
North West
East Midlands
Wales
East
West Midlands
London
South East
South West
Share of sectors with lower productivity growth in 2010–15 relative to 2000–051
%
United Kingdom 83
United States 82
France 69
Germany 62
Spain 36
1 All countries are based on 29 sectors except for Spain and the United States, which have 28. Spain does not have specific data for “other service activities”,
and the United States does not have specific data for “telecommunications”. “Activities of households as employers; undifferentiated goods- and services-
producing activities of households for own use” is excluded from calculations using KLEMS in this report because of incomplete coverage for countries in our
sample.
SOURCE: EU KLEMS (2017 release); ONS; McKinsey Global Institute analysis
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 9THE FINANCIAL SERVICES AND MANUFACTURING SECTORS AND DECLINING
TFP GROWTH PLAYED AN OUTSIZE ROLE IN THE UK SLOWDOWN
Although the slowdown was broad-based, finance and manufacturing had an outsize
impact relative to their share of the economy in the United Kingdom (Exhibit 4). Despite
making up less than 20 percent of UK value added and employment, the decline in
productivity growth in these sectors combined accounted for nearly half of the productivity
growth slowdown.11 This reflects the fact that these two sectors were the largest
contributors to a wave of particularly strong productivity growth pre-crisis and saw a
particularly dramatic slowdown post-crisis. We look at both sectors in more detail in the
next chapter.
Labour productivity growth can be decomposed into four factors: capital intensity growth,
labour quality growth, total factor productivity growth, and the mix shift effect. Capital
intensity is an indicator of access to machinery, tools, and equipment per hour worked.
Labour quality measures the impact of changes in factors like education on the productivity
of workers.12 TFP measures the efficiency with which inputs including labour, capital, energy,
materials, and purchased services are combined to produce outputs; it is often used
as a proxy for technological progress and innovation. Mix shift measures the impact on
productivity of changes in the allocation of labour across sectors with different productivity
levels.13 Analysing UK productivity growth through this lens, we find that declining TFP
growth was a discernible drag that either did not occur at all in other countries or did not
occur to the same extent (Exhibit 5).14
11
The productivity growth slowdown can be disaggregated into two components: slowdown in productivity
growth within individual sectors (a “within” effect), and the impact of reallocation of labour across sectors
with different productivity levels and shifts in relative price deflators (a “mix” effect). For the United Kingdom,
the total slowdown was 1.9 percentage points, of which 1.4 percentage points was due to within-sector
productivity growth slowdown and 0.5 percentage point to the mix effect. The within-sector effect from
financial services was a decline of 0.4 percentage point and from manufacturing 0.5 percentage point. The
total of these two, a decline of 0.9 percentage point, made up 47 percent of the total slowdown.
12
Labour quality is typically measured in growth accounting frameworks through the impact of changes in
education levels, age, and gender on the efficiency of hours worked. Growth in labour services is calculated
as the growth rate of each demographic group, weighted by its share of total wages. Labour quality growth is
the difference between growth in labour services and growth in hours worked. For additional details, see EU
KLEMS methodological materials.
13
This analysis is based on the Solow growth accounting framework using data from EU KLEMS. Note that this
decomposition is a technical accounting, with capital intensity growth reflecting increases in capital relative to
labour, labour quality growth reflecting improvements in labour (for example, through skilling), and total factor
productivity growth calculated as the residual once these effects are accounted for. The contribution of capital
intensity growth (i.e., capital services per hour worked growth) on labour productivity growth is weighted by
the capital share of total income. We have also calculated the contribution from productivity growth of each
sector (a “within” effect, which weights the contribution of a sector’s labour productivity growth by its share of
nominal GDP) and the impact of labour and relative price movements across sectors with different productivity
levels (a “mix effect”). This was done using the Generalized Exact Additive Decomposition methodology. See
Jianmin Tang and Weimin Wang, “Sources of aggregate labour productivity growth in Canada and the United
States”, Canadian Journal of Economics, May 2004, Volume 37, Number 2, and Ricardo de Avillez, “Sectoral
contributions to labour productivity growth in Canada: Does the choice of decomposition formula matter?”
International Productivity Monitor, fall 2012, Number 24. For further details on capital services, see Measuring
productivity—OECD manual: Measurement of aggregate and industry-level productivity growth, OECD, 2001.
14
We also find a substantial negative contribution of –0.5 percent coming from the mix shift effect, namely, the
movement of labour from high- to low-productivity sectors. Although there is a long-run trend of employment
moving towards low-productivity sectors such as healthcare and education, this trend does not explain the
recently observed mix effect. The –0.5 percent is higher than the effect found by other research and explained
by the inclusion of the real estate sector in our methodology. The real estate sector has a disproportionately
high productivity level as measured in national accounts using the imputed rent approach—in the United
Kingdom, for example, it had more than seven times the productivity level of the total economy in 2015. In the
2000–05 period, labour shifted fast towards the real estate sector, while the pace slowed post-crisis. Given
the high measured productivity level of the sector, small changes in the rate of labour movement to or from the
sector yield large mix shift effects, like the one we observe. For more about imputed rents and the real estate
sector, see also Rebecca Riley, Ana Rincon-Aznar, and Lea Samek, “Below the aggregate: A sectoral account
of the UK productivity puzzle,” Economic Statistics Centre of Excellence discussion paper number 2018-06.
10 McKinsey Global Institute 1. An international comparisonExhibit 4
The most important sectors contributing to the United Kingdom’s productivity growth decline were manufacturing,
financial services, and information and communication services.
Sector contribution to productivity growth difference < -0.2 -0.2 to 0 0 to 0.2 > 0.2
between 2000–05 and 2010–151
Percentage points
United United
France Germany Spain Kingdom States
Primary
Agriculture, forestry and fishing 0.0 -0.1 0.1 0.0 -0.1
Mining and quarrying 0.0 0.0 0.0 -0.1 0.2
Electricity, gas and water supply 0.0 -0.1 -0.1 -0.1 0.0
Manufacturing
Total manufacturing -0.3 -0.4 0.1 -0.5 -0.9
Boom/bust
Financial and insurance activities 0.0 0.3 -0.4 -0.4 -0.2
Real estate activities 0.1 -0.3 0.4 0.3 -0.2
Construction -0.1 0.0 0.5 0.0 0.0
Consumer-
facing Wholesale and retail trade; repair of motor
0.2 -0.2 0.2 -0.1 -0.3
services vehicles and motorcycles
Transportation and storage 0.0 -0.2 0.3 -0.1 -0.1
Accommodation and food service activities 0.0 0.0 0.2 -0.1 0.0
High-skilled
Information and communication services -0.2 0.1 -0.1 -0.3 -0.5
and support
services
Professional, scientific, technical,
0.0 0.1 0.5 -0.1 -0.1
administrative and support service activities
Other private
services Arts, entertainment, recreation, and other
-0.1 0.0 0.0 0.0 0.1
service activities
Public or
quasi-public Public administration and defence;
-0.1 0.0 -0.1 0.0 0.0
services compulsory social security
Education 0.1 0.0 -0.1 0.0 0.0
Health and social work 0.0 -0.1 0.0 0.0 0.0
Total contribution of the productivity growth decline in
-0.4 -0.7 1.3 -1.4 -2.1
each sector
Total contribution of the mix shift across sectors -0.1 -0.1 -0.2 -0.5 -0.3
Total economy productivity-growth decline -0.5 -0.8 1.1 -1.9 -2.4
1 Based on the contribution of the productivity growth decline in each sector.
SOURCE: EU KLEMS (2017 release); McKinsey Global Institute analysis
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 11Exhibit 5
Productivity growth declined 1.9 percentage points in the United Kingdom, with total factor productivity playing a
large role in the decline.
Growth rates and contributions by country Low or no effect Moderate effect Large effect
Percentage points
Decline in
productivity growth
United United
States Kingdom Germany France Spain
Productivity Change in growth, 2010–15 vs 2000–05 -2.4 -1.9 -0.8 -0.5 1.1
Growth, 2000–05 2.2 2.0 1.6 1.4 0.0
Growth, 2010–15 -0.2 0.2 0.8 0.8 1.1
Gross value Change in growth, 2010–15 vs 2000–05 -0.6 -0.8 0.8 -0.6 -3.3
added
Growth, 2000–05 2.2 2.7 0.7 1.6 3.1
Growth, 2010–15 1.6 1.9 1.5 1.0 -0.2
Hours worked Change in growth, 2010–15 vs 2000–05 1.8 1.0 1.5 -0.1 -4.4
Growth, 2000–05 0.0 0.7 -0.9 0.2 3.1
Growth, 2010–15 1.8 1.7 0.7 0.2 -1.3
Contribution Capital intensity -1.3 -0.4 -0.7 -0.5 0.5
of factors to
change in Labour quality -0.1 0.1 -0.2 0.2 0.3
productivity
growth
Total factor productivity -0.7 -1.1 0.3 -0.1 0.5
Sector mix effect -0.3 -0.5 -0.1 -0.1 -0.2
Top sectors Arts, entertainment, recreation, and other services
contributing to
the decline in Electricity, gas and water supply
productivity
growth
Financial and insurance activities
Information and communication services
Real estate activities
Manufacturing
Transportation and storage
Retail and wholesale trade
NOTE: Figures may not sum because of rounding.
SOURCE: EU KLEMS (2017 release); McKinsey Global Institute analysis
12 McKinsey Global Institute 1. An international comparisonOut of the 1.9 percentage point decline in the UK productivity growth rate, 1.1 percentage
points were due to declining TFP growth, while declining capital intensity growth was
responsible for 0.4 percentage point.15 The decline in productivity growth in both the
financial and manufacturing sectors, together with ICT services (which suffers from some
measurement issues), underpin the more significant role of declining TFP growth in the
United Kingdom compared with other countries. We calculate that the financial sector
accounted for about one third of the total decline in TFP growth, manufacturing a quarter,
and information and communications another quarter. We return to the causes of these
patterns in the financial and manufacturing sectors in the next chapter.
2. REASONS FOR THE UK SLOWDOWN
While most countries in our sample experienced a productivity growth decline across
sectors, associated with declining capital intensity and TFP growth, the United Kingdom’s
broader-based and more extreme decline in labour productivity and TFP growth indicate
that the underlying dynamic is somewhat different. In this chapter we outline that dynamic
to explain why productivity growth slowed more sharply in the United Kingdom than in other
European countries.
To start, the impact of the boom/bust cycle in the financial sector was far more significant,
accounting for about 20 percent of the aggregate productivity growth slowdown in the
United Kingdom. Then, in the aftermath of the crisis, the extent of employment growth
across regions and sectors in the United Kingdom was well ahead of European peers,
with companies on average prioritising additional labour over investment to meet demand.
Finally, while digitisation is underway across countries, with significant potential to boost
productivity growth, it is happening unevenly across industries and firms, and it comes with
adoption barriers, lags, and transition costs. The United Kingdom is no exception. While the
United Kingdom does well in internet access, basic digital skills, and the adoption of cloud
computing, it ranks poorly in other areas, especially the integration of information systems
across the value chain and across entire business processes, as well as the adoption
of robotics.
THE BOOM/BUST CYCLE IN FINANCE PLAYED A MORE SIGNIFICANT ROLE IN
THE UK PRODUCTIVITY GROWTH SLOWDOWN THAN IN EUROPE OR THE US
The impact of the boom/bust cycle in the financial sector on the UK slowdown was far
more significant than in other countries. Annual financial sector productivity growth slowed
6.1 percentage points in the United Kingdom in the post-crisis period compared with the
pre-crisis period, more than double the slowdown in the US financial sector (Exhibit 6). The
decline in productivity growth in the financial sector accounted for about 20 percent of the
aggregate productivity growth slowdown in the United Kingdom.16 Even in the United States,
which also experienced a significant boom/bust cycle, the financial sector accounted for
only about 10 percent of the productivity growth slowdown. This outsize impact cannot
be explained by the size of the financial sector compared to the rest of the economy. In the
United Kingdom, finance represents 6.9 percent of the total economy, while in the United
States, it represents 7.4 percent.
15
EU KLEMS (2017 release) data contain some internal inconsistencies to do with capital intensity figures
across different parts of the database. Some of the figures are inconsistent with other databases such as The
Conference Board and the ONS. Hence, we have slightly adjusted EU KLEMS figures in order to make them
internally consistent, which in turn brings them closer to the other databases. The changes are not material
and only slightly affect the magnitude of capital intensity, TFP, and labour quality estimates. The direction of
these estimates is consistent across databases, including EU KLEMS.
16
Financial-sector productivity growth declined by 0.4 percentage point, while the aggregate productivity
growth slowdown was 1.9 percentage points. As shown in Exhibit 4, the aggregate consists of two effects:
the total sum of the individual contributions of all the sectors, which accounts for 1.4 percentage points, and
labour movements across sectors, which account for 0.5 percentage point.
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 13Exhibit 6
The productivity growth slowdown in the United Kingdom’s financial sector was more severe than in most
other countries.
Financial sector
Productivity compound annual growth rate
%
Change in productivity growth rate
Pre-crisis, Post-crisis, relative to pre-crisis
2000–05 2010–15 Percentage points
Spain 6.8 -2.8 -9.6
United
5.4 -0.7 -6.1
Kingdom
United
2.7 -0.1 -2.8
States
France 0.4 0.9 0.5
Germany -4.2 1.8 6.0
SOURCE: EU KLEMS (2017 release); McKinsey Global Institute analysis
The outsize impact of finance in the case of the UK productivity growth slowdown is partly
due to the extent of the boom ahead
pre postof the financial crisis. In change
the period
- 2000 to 2005,
productivity growth in the financial sector in the United Kingdom was on average 5.4 percent
change +
a year, compared with 2 percent in the economy as a whole. This boost was associated with
accelerating value-added growth as loan and deposit volumes grew, helped by leverage.
Productivity growth in finance during this time was the strongest out of all UK sectors and
the highest among peers, except Spain. Then came the crisis, and with it a significant
demand shock to the financial sector, leading to a large drop in value-added growth, while
hours growth could not be readily adjusted to match. The result was a severe decline in
productivity growth to an annual average of –0.7 percent a year in the 2010 to 2015 period,
a drop of 6.1 percentage points. While it is likely that measurement issues have played some
role in this slowdown, there is also clear evidence of a decline in volume growth (see Box 3,
“What drives measured value added in the financial sector?”).
The United Kingdom experienced the second-highest increase in loan and deposit volumes
in the run-up to the financial crisis out of its peers, behind only Spain (Exhibit 7). It also
experienced the longest sustained increase in bank leverage. Yet hours worked in the sector
did not grow commensurately: only 2 percent in the United Kingdom over the period 2000 to
2007 compared to 8 percent in the United States over the same period. While technology is
likely to have contributed to banks’ efficiency, banks can also increase output in other ways
without materially increasing labour hours. For example, larger loans or higher loan-to-value
ratios translate into higher value added and do not necessarily involve more hours worked.17
17
The average loan-to-value ratio was 89 percent in 2000–05, 81 percent in 2010–15, and as low as 75 percent
in 2009. This means that, for example, for the same property (assuming constant prices), a bank would give a
smaller loan post-crisis.
14 McKinsey Global Institute 1. An international comparisonBox 3. What drives measured value added in the financial sector?
The measurement of value added is a crucial part of calculating productivity statistics. For
many sectors, it is relatively straightforward to define what goods or services are produced,
what price consumers pay for them, and how much of the change in prices is due to inflation
versus quality changes. Multiplying the amount of goods and services by the respective
deflated prices then produces a measure of real (inflation-adjusted) value added. Financial
services, however, is a more challenging sector to measure in this regard.
Financial institutions offer a range of services such as payment settlement, risk transfer,
advisory services, and intermediation. Some are relatively easy to quantify, since financial
institutions charge an explicit price or fee for them. That fee is treated like the price of any
other good or service, and hence can be used to calculate value added. Others are harder
to measure. For example, financial intermediation (taking deposits from some customers
and lending this money forward to other customers) accounts for around 40 percent of
banking gross value added. In this case, when a customer deposits her savings in a bank,
the bank does not charge an explicit fee. However, it is providing various services to that
customer, the most obvious being safekeeping. Similarly, while a bank may charge a fee
for extending a loan to a customer, a larger proportion of the profits from its loan book are
typically derived from charging interest on the loans, which partly account for the service of
connecting lenders and borrowers the bank is providing.
Measured value added attempts to size these intermediation services. Consequently, it
fluctuates with the volume of intermediation services, namely, of loans and deposits, a point
also made by the Bank of England.1 Additionally, many of the services financial institutions
charge for are directly or indirectly related to borrowing and lending, such as credit cards.
For all these reasons, the boom/bust pattern observed in the financial sector is likely to be
largely explained by the increase and decrease of loan and deposit volumes and related
dynamics in the volume of other banking services.
The Bank of England, however, argues that the contribution of finance to GDP could
be exaggerated by some challenges intrinsic to the most common technique used in
national accounts to measure value added of intermediation services, known as financial
intermediation services indirectly measured, or FISIM. The Bank of England points to various
challenges such as choosing the reference rate necessary for the calculations, deflating
variables, and measuring quality changes in banking services.2 It also finds that measured
value-added growth in finance could be influenced by changes in risk-taking.3 This raises
questions about whether the pre-crisis boom in the financial sector in the United Kingdom
might have been inflated by mismeasurement.
The Bank of England calculates that actual FISIM in the Euro-area countries, when adjusted
for risk-taking, may have been up to 40 percent lower in the 2003 to 2007 period than the
FISIM recorded in national accounts. Although this seems large, FISIM itself represents only
40 percent of value added in the banking sector, and the banking sector in turn does not
represent the entire financial sector, but around 60 percent of it. This means that errors in
measurement, while likely to be present, are not large enough to explain the entire slowdown
in productivity growth in the sector.
1
Silvana Tenreyro, The fall in productivity growth: Causes and implications, Bank of England, 2018.
2
See Measuring financial sector output and its contribution to UK GDP, Stephen Burgess, Bank of England,
2011, for percentages of different components of gross value added in the banking industry and a more
detailed explanation of some of the concepts and measurement challenges outlined here.
3
Andrew Haldane, Simon Brennan, and Vasilieos Madouros, “What is the contribution of the financial
sector: Miracle or mirage?” in The Future of Finance: The LSE Report, London, England: London School of
Economics and Political Science, 2010.
McKinsey Global Institute Solving the United Kingdom’s productivity puzzle in the digital age 15Exhibit 7
The United Kingdom’s financial sector experienced a sharp boom and bust in loan and deposit volumes, while
financial institutions did not adjust hours worked proportionately.
Compound annual growth rate 2000–05 2010–15
%
Loan volumes1 Deposit volumes1 Hours worked
6.1 3.8 1.4
France
1.9 4.5 0.9
United 10.2 8.1 1.1
States 1.6 6.1 1.3
1.1 3.2 -1.2
Germany
0.8 1.9 -0.5
United 12.2 8.2 0
Kingdom -0.5 0.7 -0.2
16.3 14.4 0.8
Spain
-5.3 -3.5 -2.6
1 Local nominal currency units.
NOTE: Sorted by growth in loan volumes in 2010–15.
SOURCE: EU KLEMS (2017 release); McKinsey Panorama; McKinsey Global Institute analysis
After the crisis, when loan and deposit volume growth fell, UK banks could not readily
reduce hours worked to match that decline. For example, loan volume growth dropped
from a rate of 12 percent pre-crisis to –1 percent post-crisis, but from 2010 to 2015, UK
banks reduced hours worked by only 0.2 percent a year.18 High fixed costs in banking, such
as IT infrastructure and branch networks, make it hard to reduce staff quickly in the face
of declining loan and deposit volumes. The cost-to-income ratio in UK financial services
increased from 63 percent in 2005 to 70 percent in 2016, driven mainly by increases in
non-staff operating costs, while growth in income remained relatively stagnant, particularly
between 2010 and 2015.19 In addition, banks needed to add staff in response to regulatory
changes, particularly in areas involving risk and compliance.20
18
UK financial firms did cut employment immediately following the crisis, although the reduction was less than
that seen in the United States. Hours worked decreased 3 percent in the United Kingdom in 2007 to 2010
while it dropped 7 percent in the United States during the same period.
19
Monetary financial institutions’ annual profit and loss, Bank of England, March 2018.
20
Martin Arnold, “HSBC wrestles with soaring cost of compliance”, Financial Times, August 4, 2014; Tom
Braithwaite, “Bubble could burst on boom in bank compliance units”, Financial Times, April 6, 2015; Laura
Noonan, “Banks face pushback over surging compliance and regulatory costs”, Financial Times, May 28,
2015.
16 McKinsey Global Institute 2. Reasons for the UK slowdownYou can also read
